There has been a recent push of research on Transformer-based models for long-term time series forecasting, even though they are inherently difficult to interpret and explain. While there is a large body of work on interpretability methods for various domains and architectures, the interpretability of Transformer-based forecasting models remains largely unexplored. To address this gap, we develop a framework based on Concept Bottleneck Models to enforce interpretability of time series Transformers. We modify the training objective to encourage a model to develop representations similar to predefined interpretable concepts. In our experiments, we enforce similarity using Centered Kernel Alignment, and the predefined concepts include time features and an interpretable, autoregressive surrogate model (AR). We apply the framework to the Autoformer model, and present an in-depth analysis for a variety of benchmark tasks. We find that the model performance remains mostly unaffected, while the model shows much improved interpretability. Additionally, interpretable concepts become local, which makes the trained model easily intervenable. As a proof of concept, we demonstrate a successful intervention in the scenario of a time shift in the data, which eliminates the need to retrain.

Recent work in XAI for eye tracking data has evaluated the suitability of feature attribution methods to explain the output of deep neural sequence models for the task of oculomotric biometric identification. These methods provide saliency maps to highlight important input features of a specific eye gaze sequence. However, to date, its localization analysis has been lacking a quantitative approach across entire datasets. In this work, we employ established gaze event detection algorithms for fixations and saccades and quantitatively evaluate the impact of these events by determining their concept influence. Input features that belong to saccades are shown to be substantially more important than features that belong to fixations. By dissecting saccade events into sub-events, we are able to show that gaze samples that are close to the saccadic peak velocity are most influential. We further investigate the effect of event properties like saccadic amplitude or fixational dispersion on the resulting concept influence.

Artificial intelligence and machine learning algorithms have become ubiquitous. Although they offer a wide range of benefits, their adoption in decision-critical fields is limited by their lack of interpretability, particularly with textual data. Moreover, with more data available than ever before, it has become increasingly important to explain automated predictions. Generally, users find it difficult to understand the underlying computational processes and interact with the models, especially when the models fail to generate the outcomes or explanations, or both, correctly. This problem highlights the growing need for users to better understand the models' inner workings and gain control over their actions. This dissertation focuses on two fundamental challenges of addressing this need. The first involves explanation generation: inferring high-quality explanations from text documents in a scalable and data-driven manner. The second challenge consists in making explanations actionable, and we refer to it as critiquing. This dissertation examines two important applications in natural language processing and recommendation tasks. Overall, we demonstrate that interpretability does not come at the cost of reduced performance in two consequential applications. Our framework is applicable to other fields as well. This dissertation presents an effective means of closing the gap between promise and practice in artificial intelligence.

Interpretable representations are the backbone of many black-box explainers. They translate the low-level data representation necessary for good predictive performance into high-level human-intelligible concepts used to convey the explanation. Notably, the explanation type and its cognitive complexity are directly controlled by the interpretable representation, allowing to target a particular audience and use case. However, many explainers that rely on interpretable representations overlook their merit and fall back on default solutions, which may introduce implicit assumptions, thereby degrading the explanatory power of such techniques. To address this problem, we study properties of interpretable representations that encode presence and absence of human-comprehensible concepts. We show how they are operationalised for tabular, image and text data, discussing their strengths and weaknesses. Finally, we analyse their explanatory properties in the context of tabular data, where a linear model is used to quantify the importance of interpretable concepts.